Single-cell phenotyping for therapeutic stratification in pancreatic cancer Pancreatic ductal adenocarcinoma (PDAC) is one of the most devastating malignancies characterized by extensive local invasion, early systemic dissemination, and pronounced resistance to chemotherapy and radiotherapy. Outcome for patients with advanced PDAC is poor, with limited clinical benefits seen with currently available therapy. Currently, clinicians and researchers cannot identify patients who are likely to progress rapidly following resection compared to those who might survive for up to 5 years. Two decades ago, Hopkins investigators identified phenotypic features, such as nuclear morphometry, in prostate cancer specimens as a modality for prognostication. Surprisingly, the classification of prostate cancers into favorable versus aggressive using morphometric features was superior to the widely used Gleason grading. Due to the low throughput nature of this methodology, however, its application to the clinical arena was impaired. Nonetheless, these studies laid the groundwork for developing a high throughput automated platform high-throughput cell phenotyping htCP assay, which can assess cellular phenotypes and ascribe accurate predictive and prognostic parameters to human cancer specimens. The aims of the project are: Aim 1: Apply htCP analysis to a panel of genetically characterized and clinically annotated PDAC cell lines to establish phenotypic signatures that correlate with the underlying mutational profile, as well as PDAC stage, overall survival, response to adjuvant gemcitabine, and pattern of subsequent metastatic failure (local recurrence vs. distant metastasis) in the corresponding patient from whom each line is derived. Aim 2: We will assess existing patient-derived xenografts in the Hopkins PDAC XenoBank and determine phenotypic signatures that correlate with PDAC stage, response to chemotherapy, pattern of subsequent metastatic failure and SMAD4/DPC4 mutational status. Using phenotypic signatures of therapy response to broad classes of anticancer agents, we will stratify the xenografts into specific treatment categories and test the htCP-predicted regimen versus gemcitabine. Aim 3: We will apply htCP analysis to primary patient tissues obtained from surgical resections. We will determine phenotypic patterns predicting the response to adjuvant chemotherapy and subsequent pattern of metastatic failure, to identify tissue-based phenotypic patterns of an aggressive vs. relatively indolent post-operative course. Aim 4: We will extend htCP analysis to formalin-fixed paraffin-embedded (FFPE) histological sections obtained from the surgical pathology archives of the Johns Hopkins Hospital. The initial tissue studies will focus on two extremes of post-operative clinical outcome- very long-term survivors vs. short-term survivors. This analysis will then be extended to a large panel of resected archival PDAC of variable survival periods, treatment response and patterns of metastatic failure in order to validate the htCP platform in patient material.